Impact of Reduced Mammographic Compression on Craniocaudal (CC) Image Quality and Average Glandular Dose (AGD): A Paired Retrospective Study

Purpose: To assess the impact of reduced mammographic compression force (10–14 daN) on craniocaudal (CC) image quality using quantitative and qualitative assessments, and to compare average glandular dose (AGD) between reduced and standard compression force (15–19 daN) in a paired retrospective cohort.

Methods/Materials: Retrospective paired analysis of 20 women (mean age, 45.6±2.6 years; BI-RADS density C/D, 95%) who underwent two CC FFDM examinations at separate visits. Fixed 8-mm circular ROIs were placed in primary fibroglandular tissue, background adipose tissue, and auxiliary fibroglandular tissue to calculate signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and a glandular visibility index (GVI) (Image 1). Three readers independently scored skin line/breast border sharpness, fibroglandular pattern visibility, and motion artifacts on a standardized 3-point scale. Interobserver agreement was evaluated using two-way random-effects ICC(2,1). SCF vs RCF metrics were compared using paired two-tailed t-tests (p < 0.05).

Results: Compression force decreased from 17.3±0.9 to 12.3±1.0 daN (−28.9%). Compressed breast thickness increased from 42.1±11.8 to 49.6±11.3 mm (+17.8%), with AGD increasing from 1.8±0.5 to 2.1±0.6 mGy (+16.7%). Quantitative metrics showed no significant differences (all p>0.05) with small effect sizes: SNR (right: p=0.083, d_z=−0.41; left: p=0.63, d_z=−0.11), CNR (right: p=0.083, d_z=−0.41; left: p=0.76, d_z=−0.07), and GVI (right: p=0.34, d_z=−0.22; left: p=0.60, d_z=0.12). Qualitative score differences were < 0.1 points for each item; reliability was good (ICC range 0.74–0.87; overall 0.80) (Figure 1).

Conclusions: In a predominantly dense-breast cohort, a 28.9% reduction in compression force maintained quantitative and reader-based CC image quality, despite a modest AGD increase attributable to greater compressed thickness. These findings support monitored compression reduction with routine tracking of thickness and AGD. Dense-breast predominance and CC-only design may limit generalizability; results characterize the dose trade-off associated with clinically feasible force reduction. Larger cohorts including MLO views are needed.